Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

Linux kernel source tree

Offline optimization of your disaggregated Dynamo graph

Machine Learning Engineering Open Book

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

Nano vLLM

High Performance LLM Inference Operator Library

[ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding

Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.

An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

A hyperparameter optimization framework

分享AI Infra知识&代码练习：PyTorch/vLLM/SGLang框架入门⚡️、性能加速🚀、大模型基础🧠、AI软硬件🔧等

A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks …

Model compression toolkit engineered for enhanced usability, comprehensiveness, and efficiency.

A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM

Provide with pre-build flash-attention package wheels on Linux and Windows platforms using GitHub Actions

Extending eBPF Programmability and Observability to GPUs (merged into https://github.com/eunomia-bpf/bpftime)

eBPF Developer Tutorial: Learning eBPF Step by Step with Examples

ebpf-go is a pure-Go library to read, modify and load eBPF programs and attach them to various hooks in the Linux kernel.

在常规推荐系统算法和系统双优化的范式下，一线公司针对单个任务或单个业务的效果挖掘几乎达到极限。从2019年我们开始关注多种信息的萃取融合，提出了OneRec算法，希望通过平台或外部各种各样的信息来进行知识集成，打破数据孤岛，极大扩充推荐的“Extra World Knowledge”。 已实践的算法包括行为数据，内容描述，社交信息，知识图谱等。在OneRec，每种信息和整体算法的集成是可插拔…

[Pytorch] Generative retrieval model using semantic IDs from "Recommender Systems with Generative Retrieval"

Distributed Compiler based on Triton for Parallel Systems

Unified Collective Communication Library

Tile-Based Runtime for Ultra-Low-Latency LLM Inference